Methodology
ENSM + MAPPO framework
Learn more →Skip to main contentFull citation & BibTeX →
Turkey 2050 Energy Transition
Hybrid AI Framework — Long-Term Renewable Energy Forecasting and Multi-Agent Reinforcement Learning Optimization for Turkey's Energy Transition to 2050
Seyit Önderol · Supervisor: Prof. Dr. Faik Bilgili · T.R. Erciyes University · February 2026
97.34%
Renewable electricity by 2050
$226.79B
Optimized energy investment
90.6%
CO₂ reduction
0.9799
ENSM forecast accuracy (R²)
13
RL algorithms compared
Executive Summary
🔴 The Problem
- • Turkey imports a large share of its energy.
- • Long-term transition planning is complex.
🔵 The Approach
- • Hybrid AI model combining energy forecasting and reinforcement learning.
🟢 The Findings
- • 97% renewable electricity is feasible with optimized investment.
- • 97.34% renewable, $226.79B investment, 90.6% CO₂ reduction.
Explore the Research
Research Timeline — Renewable Share (2000–2050)
Loading chart…
Key milestones
Historical: 2000: 18% · 2010: 32% · 2020: 45% · 2024: 54.2%
Projected (PPO): 2030: 77% · 2040: 91.6% · 2050: 97.34%
Target: 90% (exceeded by +7.34 pp)
Cite this research
Önderol, S. (2025). Long-term renewable energy forecasting and multi-agent reinforcement learning optimization for Turkey’s energy transition to 2050. PhD dissertation, T.R. Erciyes University.
Last updated: February 2026 · Data sources: IRENA, TEIAS, IEA, Eurostat · Methodology · Academic front matter